Communications of the ACM
Scientists Build a Neural Network Using Plastic Memristors
Scientists have created a neural network using plastic memristors.
Credit: Image courtesy of the study's authors
A group of Russian and Italian scientists have used plastic memristors to create a neural network.
Polymer-based memristors represent a promising area in the field of memristive neural networks. Polymeric memristors could potentially be used to build fundamentally new computers.
The scientists from the Kurchatov Institute, the Moscow Institute of Physics and Technology, Italy's University of Parma, Moscow State University, and Saint Petersburg State University discovered that creating even the simplest perceptron is not that easy. They note until the publication of their paper, there were no reports of any successful experiments--using organic materials. However, the team's experiments demonstrate it is possible to create very simple polyaniline-based neural networks.
They demonstrated that following about 12 tries their new memristive network is capable of performing NAND logical operations, and then also is able to learn to perform NOR operations.
The researchers think their research could be used to help create systems for machine vision, hearing, and other sensory organs, and also intelligent control systems for various devices, including autonomous robots.
From EurekAlert
View Full Article
Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA
No entries found